This could be two bugs, one for frames and one for tables, I'm not sure. *To repro: Follow the above link and take a looksee at the mostly minimal relevant HTML. *Expected result: Nova gets it mostly correct. There should be a gray box on the left for the iframe. There should be a vertical line separating left from right. The RHS should have a few lines of justified text. There should also be a right-aligned link that says 'MORE' with an arrow image. *Actual results: 1.The space alotted for the iframe is much larger than it should be. 2.All of the other items are there but they are crammed into a very skinny cell and you can only see the extreme left edge. (this is possibly the second (table)issue) *URL: this fault was discovered at www.fortunecity.com. *Build info: This bug is reproducible on MacOS, RHLinux, and WinNT with builds 1999040818

cc'ing chrisd in case this becomes a tables issue and she needs to see it.

I can't locate the above url from my home system. I'll try again tomorrow from the office. This sucks.

Moving to M8.

Reassigning frameset/iframe bugs to Eric.

claudius@netscape.com: could you attach the test page to this bug, so that we can see it from outside the Netscape intranet? Thanks.

Yikes, the HTML is frightening. I cleaned it up a bit, just to the point where things were properly nested, and the bug went away. I'm trying to narrow down why it displays different than Nav, but... no promises.

This is a table issue, not an iframe issue. I removed the iframe and replaced it with an image and the page layed out identically. I then entirely removed the image. Still layed out the same. I've made this page more well-formed (I'm sure it's still not 100%) and tried to make guesses at what the author intended. I then stripped out everything that seemed irrelevent and put in some colors to make things more visible. The page lays out significantly different in Gecko than in Nav 4.x, and IE 5: 1) The first cell (teal) is given too much horizontal space in Gecko. It asks for 150 pixels but gets significantly more (around 300). 2) The second cell (red) gets too much horizontal space in Gecko. It asks for 1 pixel but gets significantly more (around 20). 4) The third cell (yellow) gets too little horizontal space in Gecko. It asks for 449 pixels but gets significantly less (around 300) 3) The above two problems go away when the image is removed from the bottom cell (green).

Chris, this looks like a table bug.

Er, really, I do know that it goes "1, 2, 3, 4" and not "1, 2, 4, 3"... :) And, um, 3), which should be 4), should say "the above three..." Guess I was in a rush.

Marked [TESTCASE].

Marking with {colspan-width} in the interim. It is probably reasonable to consolidate these later into a couple of bugs. Bug #3977 is really the uber-bug for these issues (width distribution to colspanned columns).

Moving to M9.

Reassigning to Eric.

Chris, I don't see a problem with iframes here. Should I mark this bug a dup of 3977 and attach this test case there?

Chris, I don't see a problem with <IFRAME> here. Should I mark this a dup of 3977 and attach the test case there?

Eric, I shouldn't have reassigned it to you. I'm chaning the component and accepting it.

Spoke to karnaze during bug triage today. Moving to M11. Not an M10 blocker.

Moving to M13.

I'm not seeing the problems reported on the url or attachements. I'm not sure when this got fixed.

Claudius, I don't see this problem either. Could you check this please ?

With the Nov 16th build, I'm not seeing the problem. Marking as verified works for me.